If you eat plant-based and lift, someone has told you your protein is "incomplete." They may have even cited a score, DIAAS, as proof. The score is real. The conclusion people draw from it usually isn't. Let's fix that.
What DIAAS actually measures
DIAAS stands for Digestible Indispensable Amino Acid Score. It asks two questions about a protein source:
- Which indispensable (essential) amino acid is in shortest supply relative to human requirements?
- How much of that amino acid actually gets absorbed in the small intestine?
The lowest-scoring amino acid, the limiting amino acid, sets the score. Milk and eggs score above 1.0. Soy sits close behind. Most single plant foods land lower: wheat is limited by lysine, legumes tend to be limited by methionine, and digestibility is modestly reduced by fibre and antinutrients.
So far, so accurate. Here's where the popular interpretation goes wrong.
The score grades foods. You eat meals.
DIAAS evaluates one food in isolation, as the sole protein source. Nobody eats like that. The moment rice meets dal, or toast meets baked beans, or your post-workout shake combines pea and rice protein, the amino acid profiles complement each other: lysine-rich legumes patch the gap in grains, and methionine-rich grains patch the gap in legumes.
A protein score is a property of a food. Protein adequacy is a property of a diet.
And you don't need to combine them in the same meal, either. The body maintains a free amino acid pool; eating complementary proteins across the same day does the job.
What actually matters for muscle: three numbers
For someone training and eating plant-based, the evidence points to three practical levers:
- Total daily protein. This is the heavyweight variable. For muscle gain or retention in a fat-loss phase, aim for roughly 1.6–2.2 g per kg of bodyweight per day. Hit this, and protein quality differences shrink dramatically.
- Leucine per meal. Leucine is the amino acid that flips the switch on muscle protein synthesis. A practical per-meal target is ~2.5–3 g of leucine, which usually means 25–40 g of protein from plant sources (slightly more than you'd need from whey, because plant proteins run a little lower in leucine).
- Distribution. Three to four protein-anchored meals spread across the day beats one giant protein dinner.
Building the plate, no spreadsheet required
Some plant pairings that quietly solve the "incomplete" problem, the way half the world's traditional cuisines always have:
| Pairing | Why it works |
|---|---|
| Dal + rice or roti | Legume lysine + grain methionine, the original stack |
| Tofu or tempeh + anything | Soy is near-complete on its own |
| Hummus + wholemeal pitta | Chickpeas and wheat patch each other |
| Pea + rice protein blend | Blended powders mimic dairy's amino profile |
| Beans on toast | Yes, really. Britain accidentally nailed it |
If you take one action from this article: anchor each meal around a serving of legumes, soy, or seitan, and let grains and vegetables fill in around it.
The honest caveats
I'm not arguing plant and animal proteins are identical gram-for-gram. At low protein intakes, quality differences are measurable, which matters for elderly people with small appetites or anyone under-eating. And soy aside, most single plant foods genuinely are limited in one amino acid. The point is that for anyone eating enough total protein from varied sources, the limitation is solved by lunchtime without you thinking about it.
DIAAS is a useful tool for food scientists formulating products. As a reason to fear lentils? It's the most over-interpreted number in nutrition.
Want your protein targets calculated for your bodyweight, training and goals, with a meal structure you'll actually follow? That's literally what I do. Book a free discovery call below.